From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: linux-kernel@vger.kernel.org
Cc: mingo@kernel.org, jiangshanlai@gmail.com, dipankar@in.ibm.com,
        akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
        josh@joshtriplett.org, tglx@linutronix.de, peterz@infradead.org,
        rostedt@goodmis.org, dhowells@redhat.com, edumazet@google.com,
        fweisbec@gmail.com, oleg@redhat.com, bobby.prani@gmail.com,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Subject: [PATCH tip/core/rcu 09/40] rcu: Make rcu_all_qs() do deferred NOCB wakeups
Date: Wed, 12 Apr 2017 10:39:54 -0700
In-Reply-To: <20170412174003.GA23207@linux.vnet.ibm.com>
References: <20170412174003.GA23207@linux.vnet.ibm.com>
Message-Id: <1492018825-25634-9-git-send-email-paulmck@linux.vnet.ibm.com>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2112
Lines: 58

If a CONFIG_RCU_NOCB_CPUS kernel invokes call_rcu() with interrupts
disabled, wakeups must be deferred in order to avoid self-deadlock in
the cases where the disabled interrupts are due to scheduler locks being
held.  In this case, a flag is set and is checked on entry to extended
quiescent states (usermode, idle), on exit from the RCU_SOFTIRQ handler,
when the CPU in question goes offline, and on a subsequent invocation
of call_rcu().  However, a given CPU could avoid all of those states
for a considerable length of time.

This commit therefore allows an invocation of rcu_all_qs() to do the
wakeup.  It also makes the wakeup function clear the deferred-wakeup flag.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
---
 kernel/rcu/tree.c        | 6 ++++++
 kernel/rcu/tree_plugin.h | 3 ++-
 2 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index e8843bf0f5e1..eaa2b6d7abe2 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -522,6 +522,7 @@ EXPORT_SYMBOL_GPL(rcu_note_context_switch);
 void rcu_all_qs(void)
 {
 	unsigned long flags;
+	struct rcu_state *rsp;
 
 	barrier(); /* Avoid RCU read-side critical sections leaking down. */
 	if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) {
@@ -544,6 +545,11 @@ void rcu_all_qs(void)
 		rcu_sched_qs();
 		preempt_enable();
 	}
+	for_each_rcu_flavor(rsp) {
+		preempt_disable();
+		do_nocb_deferred_wakeup(this_cpu_ptr(rsp->rda));
+		preempt_enable();
+	}
 	this_cpu_inc(rcu_qs_ctr);
 	barrier(); /* Avoid RCU read-side critical sections leaking up. */
 }
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 5a149091d262..0e2114e8076b 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -2201,7 +2201,8 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp)
 {
 	int ndw;
 
-	if (!rcu_nocb_need_deferred_wakeup(rdp))
+	if (!rcu_nocb_need_deferred_wakeup(rdp) ||
+	    unlikely(rcu_scheduler_active != RCU_SCHEDULER_RUNNING))
 		return;
 	ndw = READ_ONCE(rdp->nocb_defer_wakeup);
 	WRITE_ONCE(rdp->nocb_defer_wakeup, RCU_NOGP_WAKE_NOT);
-- 
2.5.2